The major role of the Linkage Core will be to genetically assign families, largely, though not exclusively, derived from the Clinical Core, into the appropriate classes. It will be our intention to separate families by genetic analysis into four groups: (1) those with mutations in known genes (synuclein, parkin, tau for example); (2) those which probably have a pathogenic locus on 4p (parsed to Project 1); (3) those which have a pathogenic locus on chromosome 2p (parsed to Project 2) and (4) those which probably have an underlying genetic etiology, elsewhere in the genome. Families with mutation (s) in known genes will be identified by searching for segregation of marker genotypes in affecteds individuals (by both non-parametic haplotype and model-based linkage analysis) in regions formerly liked to parkinsonism/parkinson's disease. Sequencing known genes in probands will identify mutations. Families may be parsed into chromosome 2p and 4p as appropriate by comparison of autosomal dominant affected-only linkage simulation and observed lod scores/haplotype analysis. Further analysis of genetic heterogeneity among putative 2p and 4p linked families, haplotype analysis of cross-over data to prioritize regions for physical mapping (Projects 1&2) and ultimately linkage disequilibrium analysis within minimal "obligate" regions will be performed. Families excluded by linkage and lack of haplotype sharing among affecteds will be pooled into groups based on their apparent mode of inheritance (AD, AR X-linked, oligogenic/multifactorial), their phenotype (age of onset, clinical presentation, pathology, ethic and geographical origin) for future genetic studies. As new genomic loci/genes for parkinsonism become identified these will be tested by the mechanisms we have outlined. Concomitantly, the Clinical Core will expand the largest families (or ones that have the greatest potential information content for linkage/easiest to ascertain) both genealogically and in terms of sample collection. Power analysis by the Linkage Core will help prioritize family collection. Periodically, when families have been collected that are of sufficient size and statistical likelihood to successfully identify a new locus for Parkinson's disease, a genome search will be performed.